Anda di halaman 1dari 8

Clinical practice

Sex determination by the length of metacarpals and phalanges: X-ray study


on Egyptian population
Doaa A. El Morsi MD Lecturer
a,
*
, Adel A. Al Hawary MD Assistant Professor
b
a
Forensic Medicine & Clinical Toxicology, Faculty of Medicine, Mansoura University, AlGomhoreya street, Mansoura, Egypt
b
Anatomy & Consultant Radiologist, Faculty of Medicine, Mansoura University, Mansoura, Egypt
a r t i c l e i n f o
Article history:
Received 13 July 2011
Received in revised form
18 February 2012
Accepted 21 April 2012
Available online 22 May 2012
Keywords:
Sex determination
Metacarpals
Phalanges
Forensic Anthropology
Egyptians
a b s t r a c t
Measurements of hand bones length have been shown to be sexually dimorphic in many nationalities.
The aim of this study is to assess the accuracy of sex determination from the length of all metacarpals
and phalanges of right and left hands using X-ray radiographs and to develop a discriminant formula that
can be used in the Egyptians. One hundred Egyptians are included in the study (50 adult males and 50
adult females) in the period from December 2009 to January 2011 with mean age 31.60 9.44. Each is
subjected to X-ray radiographs on both hands. The results reveal that males have signicantly greater
mean values than females for all metacarpals and phalanges of both hands and the Egyptian population
has greater measurements in comparison to the other ones (e.g. Turkish and European Americans). In
addition there is no signicant difference between the right and the left hands in either males or females.
The correct classication reached an accuracy of 88%e94% by using both hands, while that for right hand
only is 88% and 88%e90% for the left hand only. Regarding the accuracy of each bone, the present results
revealed that 1st DP & PP and 3rd and 4th MC in the right and left hands are the best bones that can be
used in correct sex determination. It is concluded that the length of metacarpals and phalanges (espe-
cially the 1st DP & PP and 3rd and 4th MC) could be used for sex determination. The right hand could be
used as the left hand in determination of sex. Also the X-ray radiographs are good non invasive and
simple tool in the determination of sex from the hand bones. Furthermore the regression equation for
both hands and each hand separately is specic to Egyptian population and should be used after vali-
dation of the results in other ones.
2012 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved.
1. Introduction
The determination of race, sex, age and stature in the living or
dead is an imperative element of any medico-legal investigation.
Sexing of the remains is a cornerstone in forensic medicine as it
primarily narrows down the pool of possible victim matches.
1,2
Sex estimation is the classication of an individual as either
male or female. To achieve an assignation of sex, anthropologists
use biological traits that vary between both sexes. Although there
are only two biological sexes exist in humans, sex estimation of
a human skeleton remains a challenge.
3
There are variable techniques for detection of age and stature
which are sex dependent. Sexing is one of the simplest tasks in
forensic analysis when the external and internal genitalia can
directly suggest the sex of an individual. Sex differentiation
remains, however, a complex one in cases of intersex (where there
is a discrepancy between the external genitals and the internal
genitals); advanced state of putrefaction and mutilated, fragmen-
tary or skeletonized remains.
4e6
With the increasing frequency of mass disaster, natural or
otherwise, it is not uncommontonddismemberedhumanremains
and peripheral parts of the body.
2,7,8
For this purpose, anthropolo-
gists developed sexing methodology based on human bones. While
themost useful bones inthis regardarecranial andpelvic ones, these
are not always available.
9
In the instance of incomplete or damaged
skeletal remains, thesexdeterminationcanbeattemptedfromother
bones whichshowsome degree of sexual dimorphism.
10,11
As regard
sexual dimorphismin human sternum; although manubriumof the
two sexes is of almost equal length, the mesosternum is pro-
portionately longer in males than in females.
10
Anatomically the short tubular bones have some advantages
over other bones in a forensic context. The shafts of long bones
* Corresponding author. Tel.: +20 12 2387 2287.
E-mail address: doaaelmorsi@hotmail.com (D.A. El Morsi).
Contents lists available at SciVerse ScienceDirect
Journal of Forensic and Legal Medicine
j ournal homepage: www. el sevi er. com/ l ocat e/ j m
1752-928X/$ e see front matter 2012 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved.
doi:10.1016/j.jm.2012.04.020
Journal of Forensic and Legal Medicine 20 (2013) 6e13
often stay intact, but their epiphyses are prone to damage because
of the overlying fragile cancellous bone. However, the smaller long
bones of the hands often remain complete.
2,12
The hands, especially the metacarpal bones, have been
addressed for sex estimation with varying results in terms of
accuracy.
4,13,14
Previous studies have, with variable accuracy, shown
that the metacarpals are useful for the sex determination of skeletal
remnants. This variation was attributed to many factors including
racial or population variances. There have been fewer studies on
phalanges for the same purpose, although their accuracy has been
found to be longer than that of metacarpals.
15,16
Manolis et al.
13
mentioned that the accuracy of metacarpal bones in was ranged
from 72.3%e88.9%, while Eshak et al.
4
stated that the accuracy of
metacarpal bones was ranged from71.4 % to 92.9% and the accuracy
of phalanges was ranged from 50 % to 83%.
This study was designed to assess the accuracy of sex determi-
nation from the length of all metacarpals and phalanges of right
and left hands using X-ray radiographs and to develop a discrimi-
nant formula that can be used in the Egyptians.
2. Subjects & methods
2.1. Subjects
This study was conducted on 100 Egyptian subjects (3800
bones) classied to 50 adult males (their ages range from 17 to 53
years) and 50 adult females (their ages range from 17 to 65 years).
The study spanned the duration between December 2009 to
January 2011. Subjects were randomly chosen including every third
patient who required a hand radiograph among those attending the
Radiology Center in Mansoura University Hospital. Subjects were
excluded clinically but sometimes cases were excluded when
radiographs showed radiological skeletal immaturity, pathological
lesions such as congenital and developmental dysplasia, metabolic
bone diseases, recent trauma, recent surgery, bone tumors, osteo-
arthritis or arthritis (Done by the second author; consultant radi-
ologist & professor of Anatomy).
2.2. Methods
Right and left hands were radiographed in all subjects after
obtaining informed consent as approved by the Institutional
Research Ethics Committee. The subject was seated adjacent to the
X-ray table after wearing protective clothing with the forearm and
hand at and prone on the table with no lateral angulations at the
wrist. The hand was centered on the cassette with ngers slightly
apart from each other but at. Images were obtained using a small
focal point and a detail cassette. Exposures and distances were:
48 kV; 3.2 mA s; 90 cm source to image distance.
The length of all metacarpals and phalanges (proximal, middle
and distal) of all ngers of the right and left hands were measured
manually (in millimeters) on the X-ray lms by one observer (First
author). All measurements were repeated at the middle and end of
study to avoid any intra e observer error and the differences
between measurements found to be small and not statistically
signicant (data not shown). All measurements were made from
the midpoint of the base to the midpoint of the tip of all meta-
carpals and phalanges.
17
2.3. Statistical analysis
The statistical analysis of data was done by using SPSS (SPSS, Inc,
Chicago, IL) program statistical package for social science version
16. To test the normality of data distribution KeS
(KolmogoroveSmirnov) test was. All tested data revealed to be
parametric.
All results were expressed as mean standard deviation (SD),
minimum and maximumvalues. The analysis of the data was done
to test statistical signicant difference between groups. For quan-
titative data student t-test was used to compare between two
groups. Multivariate logistic regression analysis was used in order
to be able to predict sex based on values of predictor variables of
the metacarpals and phalanges. To detect the cutoff values with
highest sensitivity and specicity for each bone the Roc (Receiver
Operating Characteristics) curve was done. p was considered
signicant if 0.05.
3. Results
Descriptive analyses of age of the subjects are summarized
in Table 1. The mean SD age for male subjects is 31.86 8.87
years (ranges from 17 to 53 years) while mean SD age
for female subjects is 31.34 10.07 years (ranges from 17 to 65
years).
The summary statistics and comparison for all measurements
(minimum, maximum, mean SD) of the right and the left hand
bones (in millimeters) between both sexes are presented in Table 2.
All measurements are signicantly higher in males than females
regarding all metacarpals and phalanges of both right and left
hands.
A comparison between the right and the left hand bones in
males and females by using student t-test is represented in Table 3.
There is no signicant difference between the right and the left
hand in either males or females except for the rst metacarpal bone
in males.
Multivariate logistic regression analysis with the use of most
predictable measurements of both hand bones to determine sex is
demonstrated in Table 4. As a result 94% of the males (47/50) and
88% of the females (44/50) are correctly classied by using both
hands as shown in Table 5.
Furthermore multivariate logistic regression analysis with the
use of most predictable measurements of the right hand and the
left hand bones separately to determine sex is demonstrated in
Table 6.
While 88% of both males and females (44/50) are correctly
classied by using the right hand only and 90% of males (45/50) and
88% of females (44/50) are correctly classied by using the left hand
only Table 7.
In addition, the cut off values (in mm) and accuracy percentage
for sex determination for each bone in the right and left hand is
illustrated in Table 8.
So for determination of sex (z), the following equation is used
where z 0 for males and z 1 for females, was regressed on the
bone measurements (Lf PP1, Lf MC4, Rt PP2, Rt PP3 and Rt PP4) to
produce equations with the generalized formula. z Constant
(53.584) B (1.746) X (PP1) e0.640 X MC4 1.559 X
PP2 1.310 X PP3 0.815 X PP4.
If z > 0.5, the bone is male, and, if z < 0.5, the bone is female.
Fig. 1. Illustrates the measurements of all the metacarpals and
phalanges of the right and the left hands (in millimeters).
Table 1
Descriptive statistics of the studied subjects (n 100) by means of age.
Minimum Maximum Mean SD
Males (n 50) 17 53 31.86 8.87
Females (n 50) 17 65 31.34 10.07
Total (n 100) 17 65 31.60 9.44
D.A. El Morsi, A.A. Al Hawary / Journal of Forensic and Legal Medicine 20 (2013) 6e13 7
Table 2
Summary statistics (minimum, maximum, mean SD) and comparison for all measurements of the right and the left hand bones (in millimeters) between both sexes (n 100).
Bone Rt Fingers
1st 2nd 3rd 4th 5th
Min Max Min Max Min Max Min Max Min Max
DP Range Male 18.00 28.00 15.00 21.00 16.00 23.00 16.00 29.00 14.00 22.00
Mean SD 23.14 1.81 17.92 1.58 19.26 1.81 20.04 2.19 17.80 1.52
Range Female 17.00 24.00 14.00 27.00 15.00 28.00 15.00 23.00 12.00 19.00
Mean SD 20.32 1.46 16.44 1.92 17.26 2.02 17.78 1.68 15.72 1.42
t 8.544 4.188 5.194 5.755 7.037
P 0.000* 0.000* 0.000* 0.000* 0.000*
MP Range Male None 20.00 30.00 25.00 36.00 25.00 33.00 17.00 25.00
Mean SD 24.28 2.14 29.82 2.33 28.58 1.99 20.62 1.73
Range Female None 19.00 27.00 21.00 38.00 22.00 38.00 15.00 29.00
Mean SD 22.64 1.71 27.20 2.71 26.58 2.51 18.52 2.36
t None 4.222 5.169 4.408 5.058
P 0.000* 0.000* 0.000* 0.000*
PP Range Male 29.00 38.00 35.00 47.00 41.00 52.00 38.00 49.00 31.00 40.00
Mean SD 32.96 2.24 41.58 3.03 46.72 2.77 43.70 2.71 34.90 2.26
Range Female 25.00 33.00 35.00 52.00 37.00 48.00 37.00 44.00 28.00 36.00
Mean SD 28.98 2.06 38.92 2.97 42.76 2.59 40.44 2.13 32.12 1.97
t 9.216 4.424 7.360 6.661 6.547
P 0.000* 0.000* 0.000* 0.000* 0.000*
MC Range Male 43.00 55.00 63.00 80.00 60.00 77.00 54.00 69.00 49.00 63.00
Mean SD 48.44 2.74 71.58 3.93 68.62 3.72 60.66 3.35 56.72 3.35
Range Female 39.00 51.00 58.00 72.00 55.00 70.00 49.00 61.00 45.00 59.00
Mean SD 44.64 2.97 65.34 3.43 62.28 3.47 54.44 3.11 51.56 3.16
t 6.634 8.452 8.799 9.604 7.903
p 0.000* 0.000* 0.000* 0.000* 0.000*
Bone Lf Fingers
1st 2nd 3rd 4th 5th
Min Max Min Max Min Max Min Max Min Max
DP Range Male 14.00 27.00 15.00 22.00 16.00 23.00 16.00 23.00 14.00 21.00
Mean SD 22.68 2.10 18.08 1.58 19.26 1.71 19.82 1.74 17.80 1.57
Range Female 16.00 23.00 14.00 21.00 15.00 22.00 16.00 22.00 14.00 19.00
Mean SD 20.28 1.48 16.34 1.43 17.38 1.52 17.90 1.37 15.76 1.22
t 6.590 5.743 5.800 6.111 7.229
P 0.000* 0.000* 0.000* 0.000* 0.000*
MP Range Male None 20.00 29.00 25.00 35.00 25.00 33.00 17.00 25.00
Mean SD 24.24 2.01 29.80 2.26 28.56 1.96 20.58 1.83
Range Female None 20.00 27.00 21.00 31.00 23.00 37.00 15.00 22.00
Mean SD 22.62 1.66 27.12 2.08 26.44 2.33 18.66 1.74
t None 4.382 6.150 4.921 5.368
P 0.000* 0.000* 0.000* 0.000*
PP Range Male 29.00 38.00 36.00 48.00 41.00 52.00 39.00 50.00 30.00 40.00
Mean SD 33.00 2.18 41.78 2.67 46.98 2.79 44.02 2.72 34.94 2.23
Range Female 25.00 33.00 35.00 44.00 38.00 48.00 37.00 45.00 28.00 37.00
Mean SD 28.98 1.93 38.72 2.15 42.94 2.60 40.34 2.03 31.96 2.19
t 9.744 6.296 7.474 7.655 6.727
P 0.000* 0.000* 0.000* 0.000* 0.000*
MC Range Male 43.00 55.00 64.00 80.00 61.00 77.00 53.00 69.00 50.00 64.00
Mean SD 49.02 2.86 71.98 3.80 68.76 3.72 60.88 3.52 56.46 3.22
Range Female 39.00 51.00 57.00 72.00 56.00 70.00 50.00 60.00 44.00 57.00
Mean SD 44.60 2.72 65.30 3.18 62.18 3.29 54.68 2.75 51.36 3.04
t 7.900 9.507 9.353 9.792 8.139
P 0.000* 0.000* 0.000* 0.000* 0.000*
DP distal phalanges, MP middle phalanges, PP proximal phalanges, MC metacarpal bone. *p < 0.05 is signicant.
D
.
A
.
E
l
M
o
r
s
i
,
A
.
A
.
A
l
H
a
w
a
r
y
/
J
o
u
r
n
a
l
o
f
F
o
r
e
n
s
i
c
a
n
d
L
e
g
a
l
M
e
d
i
c
i
n
e
2
0
(
2
0
1
3
)
6
e
1
3
8
4. Discussion
Many studies have been attempting to determine sex by using
different body features such as foot shape, foot print ratio, foot and
shoe dimensions, the femoral head, the patella, long bones of the
arm and the teeth. Very few studies were done for the determi-
nation of sex from foot and hand dimensions.
2,14,18,19
The aim of this study is to assess the accuracy of sex determi-
nation from the length of all metacarpals and phalanges of right
and left hands using X-Ray radiographs and to develop a discrimi-
nant formula that can be used in the Egyptians.
The present study focused on one measurement (length) because
the other measurements (the width of the base and head) of meta-
carpals andphalanges are not accurate onroutinehandradiographs.
12
The present study revealed that the majority of measurements
were slightly longer in the right hand than the left hands in females,
while in males the majority of measurements were slightly longer
in the left hand than the right hand; but with no signicant
difference in the measurements between right and left hands in
either sex except for the rst metacarpal bone in males that could
be explained by daily activities which could affect the growth of
bones.
The present results were in agreement with those reported by
Krishan et al.
20,21
who studied the limb asymmetry and stated that
total upper extremity length, upper arm length and forearm length
were longer on the right side while hand length, total lower
extremity length and lower leg length were longer on the left side.
All the limb dimensions except for hand length showed statistically
signicant asymmetry (p < 0.01).
Furthermore, the nding of metacarpal length was in contra-
diction with that of Manolis et al.
13
in Athens population, who
stated that the right metacarpals were generally longer than those
of the left hand in both sexes with exceptions in some measure-
ments but these differences were not statistically signicant. Also
the present results were on contrary to that of McFadden and
Bracht
22
in European and African American populations, who
stated that the right hand was slightly longer than the corre-
sponding bones in the left hand, but the differences were small and
there were numerous exceptions. Also Krishan et al.
2
found that
hand and foot dimensions were signicantly higher (p < 0.001) in
men in both right and left sides. Hand length was found to be
Table 3
Comparison between the right and the left hand bones in males and females by using student t-test.
Bone Rt & Lf Fingers
1st 2nd 3rd 4th 5th
M F M F M F M F M F
DP t 1.949 0.350 1.307 0.440 0.000 0.496 0.984 1.030 0.000 0.270
P 0.057 0.728 0.197 0.662 1.000 0.622 0.330 0.308 1.000 0.789
MP t None None 0.221 0.111 0.141 0.270 0.163 1.124 0.531 0.655
P None None 0.826 0.912 0.888 0.789 0.871 0.267 0.598 0.516
PP t 0.275 0.000 1.492 0.714 1.695 1.243 2.030 0.798 0.286 1.091
P 0.785 1.000 0.142 0.478 0.096 0.220 0.048 0.429 0.776 0.281
MC t 3.584 0.230 1.750 0.207 0.765 0.538 1.244 1.014 1.500 0.889
P 0.001* 0.819 0.086 0.837 0.448 0.593 0.219 0.315 0.140 0.378
M male; F female; Rt right; Lf left; DP distal phalanges; MP middle phalanges; PP proximal phalanges; MC metacarpal. p < 0.05 is signicant*.
Table 4
Multivariate logistic regression analysis with the use of most predictable measurements of both hand bones to determine sex.
Variables B SE Wald Df Signicance Exp (B)
Step 1 Lf PP1 1.107 0.2270 23.874 1 0.0000 0.3310
Constant 34.216 6.991 23.955 1 0.0000 7.240
Step 2 Lf PP1 0.751 0.2420 9.604 1 0.0020 0.4720
Lf MC4 0.308 0.1210 6.509 1 0.0110 0.7350
Constant 40.926 7.958 26.450 1 0.0000 5.939
Step 3 Lf PP1 1.454 0.420 11.965 1 0.001 0.234
Lf MC4 0.536 0.171 9.824 1 0.002 0.585
Rt PP2 0.859 0.319 7.240 1 0.007 2.360
Constant 41.821 9.637 18.833 1 0.0000 1.455
Step 4 Lf PP1 1.644 0.4900 11.241 1 0.0010 0.1930
Lf MC4 0.496 0.1890 6.869 1 0.0090 0.6090
Rt PP2 1.596 0.5190 9.461 1 0.0020 4.931
Rt PP3 0.855 0.390 4.805 1 0.0280 0.4250
Constant 54.452 13.954 15.227 1 0.0000 4.449
Step 5 Lf PP1 1.746 0.5140 11.534 1 0.0010 0.1740
Lf MC4 0.640 0.2310 7.690 1 0.0060 0.5270
Rt PP2 1.559 0.5190 9.028 1 0.0030 4.752
Rt PP3 1.310 0.4760 7.577 1 0.0060 0.2700
Rt PP4 0.815 0.425 3.669 1 0.0550 2.259
Constant 53.584 13.645 15.422 1 0.0000 1.868
Rt right; Lf left; PP proximal phalanges; MC metacarpal; 1 rst nger; 2 second nger; 3 third nger; 4 fourth nger; SE standard error; Wald: c
2
;
df degree of freedom; p signicance; Exp exponential.
Table 5
The correct classication of sex by using the multivariate logistic regression of both
hands.
Sex Original Predicted group Accuracy %
Male Female Correct Incorrect
Males 50 47 3 94 % 6 %
Females 50 6 44 88 % 12 %
Overall predictive value is 91 %.
D.A. El Morsi, A.A. Al Hawary / Journal of Forensic and Legal Medicine 20 (2013) 6e13 9
signicantly larger on the right side in men only (t 2.455,
p 0.01).
Moreover the present results proved that either the right or left
hands could be used for sex determination. This is in disagreement
with the ndings of Smith
9
and Case and Ross
12
who stated that the
accuracy rates for the left hand was higher than the right hand.
While the studies of Alicioglu et al.
16
in Turkish population and
Eshak et al.
4
in Egyptian population; reside on the left hand only
because of the fact that the majority of populations are right
handed and therefore will be less inuenced by activity.
Moreover the present study revealed that males presented with
signicantly greater mean values than females (p < 0.05) for the
length of metacarpals and phalanges of all ngers of both hands.
The ordering of metacarpals (MC) by the length from longest to
shortest was 2 > 3 > 4 > 5 >1; the ordering for proximal phalanges
(PP) by the length was 3 > 4 > 2 > 5 > 1; the ordering of middle
phalanges (MP) by the length was 3 > 4 > 2 > 5 and lastly the
ordering of distal phalanges (DP) by the length was
1 >4 > 3 > 2 > 5. This could be explained by mechanical response
of the bone owing to the greater muscular demand of males.
As regard the metacarpal bones, the present study was in
agreement with the studies of Manolis et al.
13
in Athens and
McFadden and Bracht
22
in USA which stated that male metacarpals
were longer than those of females. Also McFadden and Bracht
22
mentioned that the ordering of metacarpals by the length from
longest to shortest was 2 > 3 > 4 > 5 > 1.
Regarding the ordering of proximal phalanges the present
results were in agreement with Garrido and Thompson
23
who
stated that the length order (from longer to shorter) of proximal
phalanges was found to be 3 > 4 > 2 > 5 > 1.
These ndings were on contrary to Alicioglu et al.
16
study on
Turkish, who mentioned that male measurements were greater
than females except for distal phalanges. Furthermore Eshak et al.
4
study on Egyptians by using computed tomography; stated that
males presented with signicantly greater mean values than
females for the length of the distal phalanges of all ngers, 1st and
3rd proximal phalanges and all metacarpals. Neither the middle nor
the 2nd, 4th or 5th proximal phalanges showed signicant differ-
ences between males and females.
Comparing between different populations, the present study
stated that the length of all metacarpals and phalanges of the left
hands for both Egyptian males and females were greater than the
measurements taken by Alicioglu et al.
16
; on a sample of Turkish
population.
Also the length of all metacarpals of both the right and left hands
of both Egyptian sexes were greater than the measurements taken
by Manolis et al.
13
; McFadden and Bracht
22
on a sample of Athens
and European American populations respectively.
While the measurements of all metacarpals of both right and left
hands of both Egyptian sexes in this study were smaller than the
measurements taken by McFadden and Bracht
22
on a sample of
African American population.
Table 6
Multivariate logistic regression analysis with the use of most predictable measurements of the right hand and left hand bones separately to determine sex.
Variables of Rt hand B SE Wald Df Signicance Exp (B)
Step 1 MC4 0.554 0.1050 28.073 1 0.0000 0.5750
Constant 31.914 6.032 27.990 1 0.0000 7.247
Step 2 PP1 0.574 0.213 7.295 1 0.007 0.563
MC4 0.338 0.125 7.341 1 0.007 0.713
Constant 37.233 7.335 25.769 1 0.000 1.479
Step 3 PP1 0.890 0.277 10.333 1 0.001 0.411
PP2 0.458 0.221 4.282 1 0.039 1.581
MC4 0.437 0.146 9.031 1 0.003 0.646
Constant 34.499 7.553 20.862 1 0.000 9.612
Step 4 DP1 0.627 0.280 5.008 1 0.025 0.534
PP1 0.933 0.311 8.986 1 0.003 0.394
PP2 0.648 0.264 6.020 1 0.014 1.911
MC4 0.324 0.150 4.670 1 0.031 0.723
Constant 35.330 8.078 19.129 1 0.000 2.206E15
Step 5 DP1 0.634 0.311 4.143 1 0.042 0.531
PP1 1.047 0.337 9.668 1 0.002 0.351
PP2 1.351 0.469 8.281 1 0.004 3.860
PP3 0.814 0.385 4.468 1 0.035 0.443
MC4 0.258 0.163 2.500 1 0.114 0.773
Constant 43.602 10.392 17.603 1 0.000 8.634
Variables of Lt hand B SE Wald Df Signicance Exp (B)
Step 1 MC4 0.561 0.105 28.798 1 0.000 0.571
Constant 32.297 6.008 28.894 1 0.000 1.062
Step 2 PP1 0.749 0.243 9.500 1 0.002 0.473
MC4 0.315 0.121 6.767 1 0.009 0.730
Constant 41.238 8.011 26.500 1 0.000 8.115
Step 3 PP1 1.189 0.342 12.076 1 0.001 0.305
PP2 0.680 0.302 5.090 1 0.024 1.975
MC4 0.511 0.167 9.412 1 0.002 0.600
Constant 39.032 8.493 21.123 1 0.000 8.941
Rt right hand; Lf left hand; PP proximal phalanges; DP distal phalanges; MC metacarpal; 1 rst nger; 2 second nger; 3 third nger; 4 fourth nger;
SE standard error; Wald: c
2
; df degree of freedom; p signicance; Exp exponential.
Table 7
The correct classication of sex by using the multivariate logistic regression of the
right hand and the left hand separately.
Sex Original Predicted group Accuracy % of right hand
Male Female Correct Incorrect
Males 50 44 6 88 % 12 %
Females 50 6 44 88 % 12 %
Sex Original Predicted group Accuracy % of left hand
Males 50 45 5 90 % 10 %
Females 50 6 44 88 % 12 %
Overall predictive value is 88 % for the right hand and 89 % for the left hand.
D.A. El Morsi, A.A. Al Hawary / Journal of Forensic and Legal Medicine 20 (2013) 6e13 10
These differences could be the result of genetic factors or
environmental ones affecting growth and development (nutrition,
physical activity, pathological conditions).
A study done on Egyptian population by Eshak et al.
4
by using CT
scan; revealed that the measurements of all metacarpals and
phalanges of the left hand of both sexes were slightly smaller than
the measurements of the present study. This small difference could
be explained by the different methods used for the measurements
and the difference in their accuracy and may be due to the different
sample size.
Furthermore Habib and Kamal
24
studied the length of phalanges
of the right and left hands in both Egyptian sexes with the exclusion
of thumb ngers because of their exibility as compared to other
ngers which are straight. The measurements of the distal
phalanges of both right and left hands in both sexes were greater
than those of the present study; while the middle and proximal
phalanges of both right and left hands of both sexes were smaller
than those of the present study. This could be explained by different
size sample and the difference in the method of measurements as
they measured the phalanges as the distance between two
phalange ridges by help of sliding caliper from the palmer side.
In addition to the previous ndings, the present study indicated
that ve bones of both hands (left 1st proximal phalanges, left 4th
metacarpal and right 2nd; 3rd and 4th proximal phalanges) were
useful bones for sexing of Egyptian population and showed sexual
dimorphismwith accuracy 91% (94 % for males and 88 % for females).
This is explained by differences in body size between both sexes.
The regression equation should be applied cautiously for different
population and time periods. Besides, the gathering of all the 19
bones is not easy and measurements consequently calculating
process is time consuming and exhausting task.
While by using the right hand only; ve bones also (1st proximal
and distal phalanges & 2nd; 3rd proximal phalanges & 4th meta-
carpal) were useful bones for sexing of Egyptian population and
showed sexual dimorphism with accuracy 88%. On the other side,
by using the left hand only; 3 bones (1st & 2nd proximal phalanges
& 4th metacarpal) were useful bones for sexing of Egyptian pop-
ulation and showed sexual dimorphismwith accuracy 89% (90 % for
males & 88 % for females).
This is in contrast to Alicioglu et al.
16
who obtained 72.7%
accuracy for males (16/22) and 90.7% for females (39/43) by using
the left hand only.
The misclassication in sex determination could be explained by
wrong classied cases (i.e. males of reduced dimensions or females
with very strong musculature) and due to individual and genetic
variances.
Regarding the accuracy of each bone, the present results
revealed that 1st DP & PP and 3rd and 4th MC are the best bones
that can be used in correct sex determination.
This is in accordance to Manolis et al.
13
in Athens population
who proved that 3rd & 4th MC had the highest percentage of
correct prediction. In addition Stojanowski
25
in New Mexico
revealed that the 4th MC had the highest accurate estimate.
Table 8
Cutoff values (in mm) and accuracy percentage for sex determination for each bone
(n 100).
Bone Cutoff values Accuracy % Total %
Rt 1st DP Male < 21.50 < Female Female: 85.7% 82.8 %
Male: 80%
Rt 2nd DP Male < 17.50 < Female Female: 63.3% 72.6 %
Male: 82%
Rt 3rd DP Male < 17.50 < Female Female: 81.6% 75.8 %
Male: 70%
Rt 4th DP Male < 18.50 < Female Female: 77.6% 75.8 %
Male: 74%
Rt 5th DP Male < 16.50 < Female Female: 85.7% 80.8 %
Male: 76%
Rt 2nd MP Male < 22.50 < Female Female: 79.6% 66.8 %
Male: 54%
Rt 3rd MP Male < 28.50 < Female Female: 71.4% 73.7 %
Male: 76%
Rt 4th MP Male < 27.50 < Female Female: 69.4% 69.7 %
Male: 70%
Rt 5th MP Male < 18.50 < Female Female: 91.8% 74.9 %
Male: 58%
Rt 1st PP Male < 30.50 < Female Female: 87.8% 83.9 %
Male: 80%
Rt 2nd PP Male < 41.50 < Female Female: 59.2% 72.6 %
Male: 86%
Rt 3rd PP Male < 44.50 < Female Female: 81.6% 77.8 %
Male: 74%
Rt 4th PP Male < 41.50 < Female Female: 77.6% 69.8 %
Male: 62%
Rt 5th PP Male < 32.50 < Female Female: 81.6% 72.8 %
Male: 64%
Rt 1st MC Male < 46.50 < Female Female: 71.4% 71.7 %
Male: 72%
Rt 2nd MC Male < 68.50 < Female Female: 77.6% 76.8 %
Male: 76%
Rt 3rd MC Male < 65.50 < Female Female: 77. 6% 81.8 %
Male: 86%
Rt 4th MC Male < 57.50 < Female Female: 83.7% 81.8 %
Male: 80%
Rt 5th MC Male < 52.50 < Female Female: 89.8% 80.9 %
Male: 72%
Lf 1st DP Male < 21.50 < Female Female: 80% 82 %
Male: 84%
Lf 2nd DP Male < 21.50 < Female Female: 84% 75 %
Male: 66%
Lf 3rd DP Male < 21.50 < Female Female: 86% 74 %
Male: 62%
Lf 4th DP Male < 21.50 < Female Female: 78% 74 %
Male: 70%
Lf 5th DP Male < 21.50 < Female Female: 80% 78 %
Male: 76%
Lf 2nd MP Male < 21.50 < Female Female: 66% 71 %
Male: 76%
Lf 3rd MP Male < 21.50 < Female Female: 72% 73 %
Male: 74%
Lf 4th MP Male < 21.50 < Female Female: 70% 72 %
Male: 74%
Lf 5th MP Male < 21.50 < Female Female: 72% 73 %
Male: 74%
Lf 1st PP Male < 21.50 < Female Female: 88% 85 %
Male: 82%
Lf 2nd PP Male < 21.50 < Female Female: 78% 73 %
Male: 68%
Lf 3rd PP Male < 21.50 < Female Female: 84% 77 %
Male: 70%
Lf 4th PP Male < 21.50 < Female Female: 74% 78 %
Male: 82%
Lf 5th PP Male < 21.50 < Female Female: 68% 72 %
Male: 76%
Lf 1st MC Male < 21.50 < Female Female: 76% 76 %
Male: 76%
Lf 2nd MC Male < 21.50 < Female Female: 82% 82 %
Male: 82%
Lf 3rd MC Male < 21.50 < Female Female: 82% 85 %
Male: 88%
Lf 4th MC Male < 21.50 < Female Female: 82% 84 %
Male: 86%
Table 8 (continued)
Bone Cutoff values Accuracy % Total %
Lf 5th MC Male < 21.50 < Female Female: 88% 79 %
Male: 70%
Rt right; DP distal phalanges; MP middle phalanges; PP proximal
phalanges; MC metacarpal; 1st rst nger; 2nd second nger; 3rd third
nger; 4th fourth nger; 5th fth nger.
Lf left; DP distal phalanges; MP middle phalanges; PP proximal phalanges;
MC metacarpal; 1st rst nger; 2nd second nger; 3rd third nger;
4th fourth nger; 5th fth nger.
D.A. El Morsi, A.A. Al Hawary / Journal of Forensic and Legal Medicine 20 (2013) 6e13 11
On the other hand, Eshak et al.
4
in Egyptian population proved
that the 2nd & 5th MC and 1st and 3rd PP had the highest accuracy.
Furthermore, Falsetti
26
in European and African American stated
that 2nd, 4th and 5th MC could provide method for sex assessment.
Moreover Shreuer and Elkington
27
in British White population
found that 1st MC had the highest degree of accuracy in identifying
sex. In consistent with the present result Barrio et al.
15
in Spanish
population found that the 2nd, 4th and 5th MC had the highest
accuracy.
These differences could be explained by different populations,
different sample size and different methods used in the study.
5. Conclusion
Results of the current study suggest that the length of meta-
carpals and phalanges of both hands (especially the 1st DP & PP and
3rd and 4th MC) could be used for sex determination with
reasonable accuracy. Also the X-ray radiographs are good non
invasive and simple tool in the determination of sex from the hand
bones. Furthermore the regression equation for both hands and
each hand separately is specic to Egyptian population and should
be used validated before use in other populations.
Conict of interest
The authors declare that there are no conicts of interest with
any institution or organization.
Funding
None declared.
Ethical approval
This work was approved by the Institutional Research Ethics
Committee of Faculty of Medicine, Mansoura University.
References
1. Kanchan T, Krishan K, Sharma A, Menezes RG. A study of correlation of hand
and foot dimensions for personal identication in mass disasters. Forensic Sci
Int 2010;199:112.e1e6.
2. Krishan K, Kanchan T, Sharma A. Sex determination from hand and foot
dimensions in a North Indian population. J Forensic Sci 2011;56:453e9.
3. Rsing FW, Graw M, Marr B. Recommendations for the forensic diagnosis of
sex and age from skeletons. Homo J Comp Hum Anat 2007;58:75e89.
4. Eshak GA, Ahmed HM, Abdel Gawad EAM. Gender determination from hand
bones length and volume using multidetector computed tomography: a study
in Egyptian people. J for Leg Med 2011;18(6):246e52.
5. Kanchan T, Krishan K. Anthropometry of hand in sex determination
of dismembered remains e a review of literature. J for Leg Med 2011;18:14e7.
6. Donohoue PA. Disorders of sex development (intersex). In: Nelson Textbook of
Pediatrics. Kliegman, R. M, Behrman, R.E, Jenson, H.B. and Stanton, B.F. (Eds.).
19th ed, Saunders Elsevier, Philadelphia, P. 582.
7. Kanchan T, Kumar GP, Menezes RG. Index and ring nger ratio e a new sex
determinant in South Indian population. Forensic Sci Int 2008;181(1):53e4.
8. Kanchan T, Kumar GP, Menezes RG, Rastogi P, Rao PPJ, Menon A, et al. Sexual
dimorphism of the index to ring nger ratio in South Indian adolescents. J for
Leg Med 2010;17:243e6.
9. Smith SL. Attribution of hand bones to sex and population groups. J Forensic Sci
1999;41:469e77.
10. Hunnargi SA, Menezes RG, Kanchan T, Lobo SW, Uysal S, Herekar NG, et al.
Sternal index: is it a reliable indicator of sex in the Maharashtrian population of
India? J for Leg Med 2009;16:56e8.
11. Lazenby RA. Identication of sex from metacarpals: effect of side asymmetry.
J Forensic Sci 1994;39(5):1188e94.
12. Case DT, Ross AH. Sex determination from hand and foot bone length. J Forensic
Sci 2007;52:264e70.
13. Manolis SK, Eliopoulos C, Koilias CG, Fox SC. Sex determination using
metacarpal biometric data from Athens collection. Forensic Sci Int
2009;193:1e6.
14. Ozden H, Balci Y, Demirustu C, Turgut A, Ertugrul M. Stature and sex estimate
using foot and shoe dimensions. Forensic Sci Int 2005;29:181e4.
15. Barrio PA, Trancho GJ, Snchez JA. Metacarpal sexual determination in
a Spanish population. J Forensic Sci 2006;51:990e5.
16. Alicioglu B, Yilmaz A, Karakas HM, Cigali BS. Sex determination by
interarticular distance of metacarpals and phalanges: a digital radiologic study
in contemporary Turkish people. J Anat 2009;3:14e20.
17. Robertson J, Zhang W, Liu JJ, Muir KR, Maciewicz RA, Doherty M. Radiographic
assessment of the index to ring nger ratio (2D: 4D) in adults. J Anat
2008;212:42e8.
18. Agnihotri AK, Purwar B, Jeebun N, Agnihotri S. Determination of sex by hand
dimensions. Internet J Forensic Sci 2006;1(2).
19. Moudgil R, Kaur R, Menezes RG, Kanchan T, Garg RK. Foot index: is it a tool for
sex determination? J for Leg Med 2008;15:223e6.
20. Krishan K, Kanchan T, DiMaggio JA. A study of limb asymmetry and its effect
on estimation of stature in forensic case work. Forensic Sci Int
2010;200:181.e1e5.
21. Krishan K. Marked limb bilateral asymmetry in an agricultural endogamous
population of North India. Am J Hum Biol 2011;23(5):674e85.
22. McFadden D, Bracht MS. Sex and race differences in the relative length of
metacarpals and metatarsals in human skeletons. Early Hum Dev
2009;85:117e24.
23. Garrido - Varas CE, Thompson TJU. Metric dimensions of the proximal
phalanges of the human hand and their relationship to side, position, and
asymmetry. HOMO - J Comp Hum Biol 2011;62:126e43.
Fig. 1. A radiograph demonstrates the measurements of all metacarpals and phalanges of the right and the left hands (MC metacarpal; PP proximal phalanges; MP middle
phalanges and DP distal phalanges).
D.A. El Morsi, A.A. Al Hawary / Journal of Forensic and Legal Medicine 20 (2013) 6e13 12
24. Habib SR, Kamal NN. Stature estimation from hand and phalanges length of
Egyptians. J for Leg Med 2010;17:156e60.
25. Stojanowski CM. Sexing potential of fragmentary and pathological meta-
carpals. Am J Phys Anthropol 1999;109:245e52.
26. Falsetti AB. Sex assessment from metacarpals of the human hand. J Forensic Sci
1995;40:774e6.
27. Scheuer JL, Elkington NM. Sex determination from metacarpals and the rst
proximal phalanx. J Forensic Sci 1993;38:769e78.
D.A. El Morsi, A.A. Al Hawary / Journal of Forensic and Legal Medicine 20 (2013) 6e13 13